A tutorial on wavelets from an electrical engineering perspective.I. Discrete wavelet techniques

The objective of this paper is to present the subject of wavelets from a filter-theory perspective, which is quite familiar to electrical engineers. Such a presentation provides both physical and mathematical insights into the problem. It is shown that taking the discrete wavelet transform of a function is equivalent to filtering it by a bank of constant-Q filters, the non-overlapping bandwidths of which differ by an octave. The discrete wavelets are presented, and a recipe is provided for generating such entities. One of the goals of this tutorial is to illustrate how the wavelet decomposition is carried out, starting from the fundamentals, and how the scaling functions and wavelets are generated from the filter-theory perspective. Examples (including image compression) are presented to illustrate the class of problems for which the discrete wavelet techniques are ideally suited. It is interesting to note that it is not necessary to generate the wavelets or the scaling functions in order to implement the discrete wavelet transform. Finally, it is shown how wavelet techniques can be used to solve operator/matrix equations. It is shown that the “orthogonal-transform property” of the discrete wavelet techniques does not hold in numerical computations     

POEMS: end-to-end performance design of large parallel adaptive computational systems

The POEMS project is creating an environment for end-to-end performance modeling of complex parallel and distributed systems, spanning the domains of application software, runtime and operating system software, and hardware architecture. Toward this end, the POEMS framework supports composition of component models from these different domains into an end-to-end system model. This composition can be specified using a generalized graph model of a parallel system, together with interface specifications that carry information about component behaviors and evaluation methods. The POEMS Specification Language compiler will generate an end-to-end system model automatically from such a specification. The components of the target system may be modeled using different modeling paradigms and at various levels of detail. Therefore, evaluation of a POEMS end-to-end system model may require a variety of evaluation tools including specialized equation solvers, queuing network solvers, and discrete event simulators. A single application representation based on static and dynamic task graphs serves as a common workload representation for all these modeling approaches. Sophisticated parallelizing compiler techniques allow this representation to be generated automatically for a given parallel program. POEMS includes a library of predefined analytical and simulation component models of the different domains and a knowledge base that describes performance properties of widely used algorithms. The paper provides an overview of the POEMS methodology and illustrates several of its key components. The modeling capabilities are demonstrated by predicting the performance of alternative configurations of Sweep3D, a benchmark for evaluating wavefront application technologies and high-performance, parallel architectures.     

Symbol error rate of selection amplify-and-forward relay systems

Cooperative diversity schemes significantly improve the performance of wireless networks by transmitting the same information through several nodes. The amplify-and-forward (AF) relaying method is one of the most attractive cooperative diversity schemes due to its low complexity. Selection AF relaying has recently been proven to achieve the same diversity order as and lower outage probability than all-participate relays. In this letter, we present an asymptotic analysis of the symbol error rates of a selection AF network, and compare it with the conventional all-participate scheme     

Simplifying Control Flow in Compiler-Generated Parallel Code

Optimizing compilers for data-parallel languages such as High Performance Fortran perform a complex sequence of transformations. However, the effects of many transformations are not independent, which makes it challenging to generate high quality code. In particular, some transformations introduce conditional control flow, while others make some conditionals unnecessary by refining program context. Eliminating unnecessary conditional control flow during compilation can reduce code size and remove a source of overhead in the generated code. This paper describes algorithms to compute symbolic constraints on the values of expressions used in control predicates and to use these constraints to identify and remove unnecessary conditional control flow. These algorithms have been implemented in the Rice dHPF compiler and we show that these algorithms are effective in reducing the number of conditionals and the overall size of generated code. Finally, we describe a synergy between control flow simplification and data-parallel code generation based on loop splitting which achieves the effects of more narrow data-parallel compiler optimizations such as vector message pipelining and the use of overlap areas.     

Rsim: simulating shared-memory multiprocessors with ILP processors

The early 1990s saw several announcements of commercial shared-memory systems using processors that aggressively exploited instruction-level parallelism (ILP), including the MIPS R10000, Hewlett-Packard PA8000, and Intel Pentium Pro. These processors could potentially reduce memory read stalls by overlapping read latency with other operations, possibly changing the nature of performance bottlenecks in the system. The authors' experience with Rsim demonstrates that modeling ILP features is important even in shared-memory multiprocessor systems. In particular, current simple processor-based approximations cannot model significant performance effects for applications exhibiting parallel read misses. Further, recent shared-memory designs such as aggressive implementations of sequential consistency use the aggressive ILP-enhancing features of modern processors that simple processor-based simulators do not model. As microprocessor systems become more complex, the availability of shared infrastructure source code is likely to become increasingly crucial. The authors plan to release a new Rsim version shortly that will include instruction caches, TLBs, multimedia extensions, simultaneous multithreading, Rabbit fast simulation mode, and ports to Linux platforms     

Guest Editor's Introduction: Reliability-Aware Microarchitecture

This special issue of IEEE Micro addresses the trends toward reliability-aware microarchitecture, their system-level implications, and some innovative ideas on how to build systems in the face of those challenges.     

Minimum norm mutual coupling compensation with applications in direction of arrival estimation

This paper introduces a new mutual coupling compensation method based on the minimum norm solution to an underdetermined system of equations. The crucial advantage over previous techniques is that the formulation is valid independent of the type of antenna element and provides good results in situations where signal strengths vary considerably. In using the matrix pencil algorithm to estimate the directions of arrival, the examples show that the proposed method results in significantly lower bias than the traditional open circuit method. The analysis of mutual coupling is also applied in the context of a code division multiple access communication system.     

Efficient solution of the differential form of Maxwell's equationsin rectangular regions

One of the problems of the finite element and the finite difference method is that as the dimension of the problem increases, the condition number of the system matrix increases as Θ(1/h² ) (of the order of h², where h is the subsection length). Through the use of a suitable basis function tailored for rectangular regions, it is shown that the growth of the condition number can be checked while still retaining the sparsity of the system matrix. This is achieved through a proper choice of entire domain basis functions. Numerical examples have been presented for efficient solution of waveguide problems with rectangular regions utilizing this approach     

Molecular communication using Brownian motion with drift

Inspired by biological communication systems, molecular communication has been proposed as a viable scheme to communicate between nano-sized devices separated by a very short distance. Here, molecules are released by the transmitter into the medium, which are then sensed by the receiver. This paper develops a preliminary version of such a communication system focusing on the release of either one or two molecules into a fluid medium with drift. We analyze the mutual information between transmitter and the receiver when information is encoded in the time of release of the molecule. Simplifying assumptions are required in order to calculate the mutual information, and theoretical results are provided to show that these calculations are upper bounds on the true mutual information. Furthermore, optimized degree distributions are provided, which suggest transmission strategies for a variety of drift velocities.     

Space-time adaptive processing using circular arrays

A direct data-domain (D³) least-squares space-time adaptive-processing (STAP) approach is presented for adaptively enhancing radar signals in a non-homogeneous environment of jammers, clutter, and thermal noise, utilizing a circular antenna array. The non-homogeneous environment may consist of non-stationary clutter. The D ³ approach is applied directly to the data collected by a circular antenna array (utilizing space), and in time (Doppler) diversity. Conventional STAP generally utilizes statistical methodologies, based on estimating a covariance matrix of the interference, using the data from various range cells of the circular array and assuming that it is a uniform linear array. However, for highly transient and inhomogeneous environments, the conventional statistical methodology may be difficult to apply. Moreover, the error in forming the covariance matrix by assuming that the data collected by the circular array is assumed to be a uniform linear array is highly problem dependent. Hence the D³ method is presented, as it analyzes the data in space and time over each range cell separately. However, it treats the antenna array as circular, i.e., it deals with the antenna structure in its proper form. Limited examples are presented to illustrate the application of this approach